Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tnzr20

Download by: [121.75.88.191] Date: 27 March 2017, At: 18:50

Journal of the Royal Society of New Zealand

ISSN: 0303-6758 (Print) 1175-8899 (Online) Journal homepage: http://www.tandfonline.com/loi/tnzr20

The prehistory of the Southern Wairarapa

B. Foss Leach

To cite this article: B. Foss Leach (1981) The prehistory of the Southern Wairarapa, Journal of theRoyal Society of New Zealand, 11:1, 11-33, DOI: 10.1080/03036758.1981.10419449

To link to this article: http://dx.doi.org/10.1080/03036758.1981.10419449

Published online: 21 Dec 2011.

Submit your article to this journal

Article views: 236

View related articles

Citing articles: 1 View citing articles

(C)Journal 0/the Royal Society 0/New Zealand,Volume 11, Number 1, pp. 11-33, J0 jigs.

The Prehistory ofthe Southern Wairarapa

B. Foss Leach"

Archaeological evidence for prehistoric settlement in the southern Wairaraparegion of New Zealand is reviewed. The widely-used dichotomy of Archaic andClassic Maori is not found to be appropriate to this region. Instead, the threads ofstability and change in economy, settlement pattern, material culture, technology,and communication patterns are placed in a setting ofthe complex environmentalhistory of the region. A series of seven cultural periods, spanning ADIIOO toAD1853, is suggested.

INTRODUCTION

The environmental setting

The southern Wairarapa (Fig. 1) consists of three different environmental zones - amountainous hinterland, a complex alluvial valley with associated rivers and lakes, and anarrow coastal platform. The Rimutaka range to the west rises to 1000 m and largelyconsists ofJurassic greywackes. This is a substantial natural barrier to communicationbetween groups on either side. In the east is the Aorangi mountain block, also about1000 m high. Geologically, the Aorangi mountains are a mosaic of Triassic greywackesand sandstones, Lower Cretaceous rocks, and Miocene mudstones. Between these twomountainous areas occurs the rich alluvial valley of the Ruamahanga River. Estuarineconditions prevailed in this valley until about 1500 B.c., when it turned into thelacustrine complex of today (Leach and Anderson, 1974). The coastal platform of PalliserBay is a narrow unstable strip about 200 m wide, with only occasional outcrops of therocky basement. The sediments are largely redeposited gravels from the Aorangimountains. They have been produced by massive erosion processes which characterizethe area and have been thrown up against the cliffs and foothills. In the Putangirua,Pinnacles, and H urupi area, the sediments are of marine Miocene origin.

Forests in the region vary greatly from swamp kahikatea stands in the alluvial valley, toa patchwork of mixed broadleaf podocarp forest and beachlands in the mountains. Theclimax vegetation of the coastal strip was a scrubland composed of flax, kanuka, karaka,ngaio, and occasional kowhai and cabbage trees.

Palliser Bay is a rich marine environment with soft unstable foreshores and occasionalrocky headlands such as Black Rocks. The rivers and lakes debouch into the bay througha narrow channel in a gravel bar which tenuously separates Lake Onoke from the sea.

The climate is variable from the occasional long summer drought through to wildstorms in winter. Two winds prevail; a northwesterly which can be either dry or wet andlast for long periods; and a southerly, which brings the main storms. Palliser Bay andCook Strait generally are well known as a tumultuous stretch of sea. Rainfall in PalliserBay is about 1000 mm, and the area has a mean monthly temperature range of about 7 to19° C over winter and summer.

Economic resources

An important area of chert occurs on the east coast of the Wairarapa near Tora, about25 km north of Cape Palliser. This varies in colour and some has light chalcedonicveining. It is generally of high quality and is a very suitable material for making knives,

... Department of Anthropology, University of Otago, Box 56, Dunedin, New Zealand.

12 Journal ofthe R oyal Society ofN ew Zealand

scrape rs and th e like. Both at T ar a , and farther south at White Rock , siliceous limestoneis found, and while it vari es a great deal in quality, some may have bee n suita b le formaking small adzes. A type of fine-grained unbaked 'a rgi llite ' occurs sporadica lly inPalliser Bay and is well suited to manufacturing small items suc h as stone lur es. The mainuseful rock in th e area, however, is greywacke. This occurs in supe r-ab undance a longbeach es and in th e int erior. It could be used for some types of adze and a lso for ovenstones.

Botanical resources of the region include a considerabl e var iety of timbers and wildvegeta ble produce. Podocarpus totara was found only in th e main alluvia l valley and would

"

I,,

I

II

/

'" v;; .NG"'1>0 Iy'r , > TE ~AU~AU POINT, - ' TIIl , e- 190 . - .

.. ' C'<\"

~ . ...1

oJ!'

,.'

..-

,.ot..

c" ", /o •

<- ~.....:F r-'/

0"Ii

Q:'>/ ;

. ,of'

.J"

II

\ ROCKS Ell _ . _ I I

BLA~K \ \,\,\S _ , u GJ.fTI16u ', 'Pi l'~ . ,.1 .-' S€'C,.

«..

,,

,\-,

I

... c, "",II

'..,

TE H.JMEN(;A

~ '0 0 ,

""

I

"

"

" ,

WHANGAIMoANA. - - P4l l IS~~"

&41',

....

NIt

I •.rI "

-- '.-, '

I,

' ... , ~r " «-

~d \7

-~---=-­

kik:metraesCONTOURS IN MET k ES

SOUTHERNWAIRARAPAo_.

41°30 ' S

;!' -/

.J"

Fig. I - Location map.

Leach - The Prehistory if the Southern Wairarapa 13

have been the main. source of timber for making canoes. The main type of edible fernroot,Pteridium esculentum, is very rare in the Aorangi mountains, but again could have beenexploited in the Ruamahanga valley. The forest of course is rich in many kinds of edibleberries such as those ofkaraka and hinau. Flaxes, sedges and rushes are readily availableand would be suitable materials for garments and cordage.

The main source of information about moa is from archaeological sites. It suggests thatmoa were either very rare or even absent in the southern Wairarapa before the time offirst settlement (Leach and Leach, 1979: 236). This is interesting, because at the end ofthe Pleistocene period there were at least three genera of moa living in the Aorangimountains, evidenced by material from the Martinborough caves - A nomalopteryx,Pachyornis, and Dinornis (Leach and Leach, 1979: 235). Other types of bird, both of theforest and sea, were numerous during the last millennium.

Cook Strait is well known for the quantity of whales and seals of various kinds whichregularly pass through it. There is a small colony of seals at Black Rocks, but the mainsource of sea mammals was probably the occasional stranding, either as single individualsor en masse. Strandings still occur today.

Man brought the Polynesian rat and dog to New Zealand, and these were also animportant economic resource to people living in the southern Wairarapa.

Finally, the marine and freshwater resources must be mentioned. Palliser Bay is rich infish, crayfish, and rocky shore shellfish. The latter only occur in isolated patches atheadlands but can be present in large numbers. Eels occur in great abundance duringseasonal mass migrations to the sea through the mouth of the gravel bar at Lake Onoke.Eels were also available in smaller numbers, along with freshwater crayfish, throughoutthe year in the smaller rivers and streams throughout the region.

Traditional occupation of the southern Wairarapa

Traditions of Maori occupation in the Wairarapa extend all the way back to the greatadventurer Kupe. His visit to this area is immortalised in two place names: nga-waka-o­Kupe, an area on the western slopes of the Aorangi mountains which looks like threeupturned canoes; and nga-ra-o-Kupe, near Cape Palliser, an upright slab of rock whichlooks like a triangular lateen sail.

At the time of European contact, the Ngati-Kahungunu were in effective control oftheWairarapa. This tribe had dominion at that time over a large area extending fromHawkes Bay to Cook Strait along the east of the main divide. Oral traditions consistentlyrefer to a mass migration of these people into the Wairarapa in the early 17th century. Atthis period, the Rangitane were in occupation, although other groups may have beenthere also. The Tini-o-awa and the Ngati Ira are known to have co-existed with theRangitane in this area before moving elsewhere.

The nature and distribution of archaeological sites

Some 249 archaeological sites have been recorded in the region; 46 of these in themain alluvial valley, 87 along the shores of Palliser Bay, and a further 116 along the eastcoast as far north as Glenburn. This is not a dense concentration (one site for each6.83 km"), by some standards in the North Island, and is partly a reflection of the lowcarrying capacity of the land, given the kind of economic system which might haveprevailed during the prehistoric period.

In the alluvial valley, a few sites occur in the Western Lakes area and the eastern valleysand dunes (thrown up during the much earlier estuarine conditions); while rathermore are found in the eastern valley lowlands and associated river valleys running ofTintothe Aorangi mountains. In Palliser Bay sites tend to be concentrated about the mouthsand inner reaches of the seven main river valleys which drain the Aorangi block. Thistype of distribution is not so apparent with sites farther up the east coast.

The most common type of site is a small midden and/or oven area, but almost ascommon are stone wall complexes along the coastal flats. These are particularly strikingfeatures (Fig. 2), and lack of ploughing along these flats has helped to preserve them.

14 Journal 0/ the Royal Society 0/New Zealand

Fig. 2 - Stone walls are common in Palliser Bay and outlined early garden areas. This example(NI68/29) is unusual in being 2 km inland at the Washpool. Gardening began here by the middleof the 13th century AD.

Fig. 3 - The rectangular raised-rim pit is a striking field feature in the southern Wairarapa. Thistype of kumara storage appears to have been imported from farther north in the 16th century AD.These examples are in the Whakatomotomo Valley (NI65/39).

Leach - The Prehistory ofthe Southern Wairarapa 15

Fig. 4 - Battery Hill Pa (NI65/18) in the Western Lake area. This site has a ring ditchreminiscent of sites in Taranaki and may well have been built during raids in the early 19thcentury AD.

Groups of storage pits are almost as numerous (Fig. 3), and occur either on the exposedcoastal flat, or as defended complexes in 'hide-away' locations on spurs in the bush. A feware defended in the manner of pa sites, with a combination of naturally steep slopes andditch and bank features.

Fortified settlements, or pa, are fairly rare (37), and two-thirds of these are along thecoast north of Cape Palliser. A particularly striking example;' known as Battery Hill(N 165/18) on a Lands and Survey Farm (Fig. 4), is defended in similar manner to somering ditch pa in the Taranaki district and may have been built by Taranaki raiders, whoare known to have been in the area about AD 1825 (Mair, 1979: 13). A similar one isnear Pirinoa across the lake on Hume's property (NI65/37). Open settlements inundefended locations number about 15, and are fairly evenly distributed. Finally, thereare a few more specialised sites such as garden soils, karaka groves (possibly planted),burial grounds, and chance finds of artefacts.

Review of archaeological researchThe first significant publication resulted from a survey by Adkin (1955; see also Leach

and Leach, 19}9b), who drew attention to the apparent high density of settlement in theeastern Palliser Bay area. The artefacts which he found were typical of moa-hunter sites,as are many items now in private hands (Budd collection, Holmes Warren collection, andBroughton collection) or in the National Museum.

In the late 1950s at least four burial sites were excavated - one near Honeycomb

16 Journal if the Royal Society ifNew Zealand

Rock which had an unusual shell tiki included (Barrow, 1959: 6), and a further three atthe mouth of the Pararaki River (Davis, 1959: 18; Keyes, 1975, pers. comm.). One ofthese was accompanied by a shark tooth necklace (Cairns, 1971). Several shortpublications (Cairns 1959 a, b, 1960, 1961; Davis, 1957; Palmer, 1961) also appearedabout this time, and the growing interest in the Wairarapa culminated in a field trip tothe Glenburn area in 1960 by participants of the New Zealand ArchaeologicalAssociation conference (Scarlett, 1960: 10).

Wellman included parts of the Wairarapa coast in his coastal reconnaissance of theNorth Island. While he drew attention to the presence of moa bone in close associationwith human occupation at Cape Palliser, he did not believe there had been significantsettlement in the region.

The coast is extremely windswept and so inhospitable that the small population itwould have supported are 'unlikely to have left a continuous record of theirpresence .... (Wellman, 1962: 39)

The systematic recording of prehistoric sites in the Wairarapa began in 1961 with Smartand Cameron's survey of the Pahaoa area (Smart, 1966). This was followed by areconnaissance of the coastline between Cape Palliser and Pahaoa in the summer of1963-64 by Hitchings and O'Rourke (n.d.). A group of students led by Mitcalfe (1968a-f; 1969) revisited parts ofthis area on a series of weekends in 1968, and also extendedthe survey farther north.

Until Park's study (1970) of an area high in the Tararu. l Ranges in 1968-69, no directindication ofthe time depth ofWairarapa prehistory was available. This excavation of anoven complex, dated to AD 1209 ± 54, yielded important palaeoclimatic information (allradiocarbon dates in this paper refer to the old half life, without secular correction).

From 1969 to 1972 a regional programme of archaeological research was centred onthe southern Wairarapa region. This included a site survey of some 1700 kms, analysis ofhistoric and traditional records, and excavations of25 archaeological sites in Palliser Bay.This project co-ordinated the efforts of some of the staff and students of the OtagoUniversity Anthropology Department, the bulk of the work being carried out by A. ].Anderson, B. F. Leach, H. M. Leach, G. M. Mair, K. Prickett, N.]. Prickett, I. W. G.Smith, D. G. Sutton, and R Wallace. Results of this work have been extensivelypublished, the main work being a bulletin of collected papers (Leach and Leach, 1979a).Since most is known about the prehistoric occupants of Palliser Bay, the followingreconstruction is most relevant to that area of the Southern Wairarapa.

THEMES IN WAIRARAPA PREHISTORYThe people

It is logical to begin a review of prehistory in a region by considering the peoplethemselves. In the Wairarapa we are fortunate that a detailed study (Sutton, 1974, 1979)has been made of skeletal material recovered in Palliser Bay. From the time of firstsettlement of the region, the people of the southern Wairarapa clearly had typicalPolynesian characteristics (Houghton, 1980). They were tall by contemporary Europeanstandards, adult males having an average height of 175 ± 2 cm, and adult females161 ± 3 ern. They lived a vigorous but fairly short life; the average age at death of adultswas about 38 years. About 70 per cent of adults died between the age of 32 and 44, andonly one in a hundred adults reached the grand old age of 56 years. This may seem anindication ofvery short life expectancy but was common in prehistoric New Zealand, andfor some populations, such as those at Wairau Bar (Houghton, 1975), life expectancy wasconsiderably shorter.

Only a small number of children were born in anyone fa-nily, about two to four beingthe norm. Most adults suffered from osteoarthritis in their limbs, and spinal degenerationwas common in advancing years. Causes of death are usually very difficult to determinearchaeologically, but it seems that these people lived active lives without much traumaticinjury such as broken limbs. There is no evidence of group warfare in the early periods.

The evidence relating to teeth is striking. This shows every sign that the diet was

Leach - The Prehistory ofthe Southern Wairarapa 17

rather coarse and abrasive. Significant tooth wear seems to have begun as early as six orseven years, and by the time a person reached 20 to 30, the front teeth and first molarswere becoming quite worn. By an age of 30 to 40 years the dentition was badly depleted,and anyone who survived into the next decade would have had marked periodontaldisease and abscessing. This reconstruction is a sorry picture of the effects of hard life inPalliser Bay, where teeth were used not only to chew gritty food but also as implementsfor biting and stripping.

Some lines of evidence suggest that in the course of the prehistoric period, the generalphysical conditions of the people deteriorated significantly. This is particularly evident inthe dense and numerous Harris lines visible in the X-rays of limb bones of both childrenand adults dated to the late 15th century. In contrast, these features are absent amongstearlier people. These Harris lines result from episodes of malnutrition or moderatelysevere illness, and show that the general conditions of life must have worsened markedlyafter occupation became established in the area. This is a most interesting line ofevidence, and fits well with the economic changes which appear to have taken place overthe same period (see below).

Prehistoric horticultureThere are several lines of evidence which show that a form of horticulture was

practised in Palliser Bay from the time of first settlement. One of the most obvious types offield feature of the region is a long row of piled up stones - 'stone walls' - or stonescarefully placed in single boulder alignments. These features are organised intocomplexes, and H. Leach (1979a: 137ff.) has shown that they are field boundaries,surrounding early garden plots. Excavations revealed these walls to have been 2 to 3 mwide, and their maximum height was probably about 40 to 60 ern. Charcoal, and evensome midden, is often associated, but in some cases there is little or no matrix other thanstones. The walls are partly related to clearing operations in the preparation of gardenplots and also functioned as boundary markers. Most are very straight, and some are aslong as 200 m.

In addition, there are pits, either associated with the walls, or in areas by themselves.These are clearly for the storage of horticultural produce. Clusters of walls are found ateach of the seven main valley systems in Palliser Bay. They have been dated to as early asthe beginning of the 12th century and as late as the close of the 15th century AD.

In some cases, cleared garden plots have numerous small mounds of stones piled upinside the boundaries, and these are thought to be small cultivation mounds (H. M.Leach, 1979b: 242; B. F. Leach, 1979a: 122). There are also a number of small terraces(B. F. Leach, 1979a: 111), of varying size, which are not demarcated by walls or stonesbut nevertheless were garden plots (see Fig. 5).

Determining what crops might have been grown in these gardens is a relatively simplematter, despite the absence of archaeological evidence in the form of actual pieces of theplants. Amongst the cultigens imported into New Zealand in the prehistoric period,kumara requires a growing season of five to six months, gourd six to seven months, yamabout eight months, and dry land taro seven to nine months. Both kumara and gourdgrow best at mean temperatures above 17° C, while taro and yam prefer a meantemperature of above 20° C during the growing season (H. M. Leach, 1979b: 242). Thetubers of kumara and yam suffer tissue damage if and when the temperature falls below 4to 5° C, while taro is rather more tolerant. The kumara can tolerate drier conditions thantaro and yam: minimum monthly rainfall for kumara is 60 mm, while taro and yamrequire at least 100 mm, preferably up to 150 mm during the growing season.

In modern Palliser Bay, the effective growing season is five to six months, with a meantemperature of ) 6.3° C. Rainfall ranges from 58 to 82 mm during the growing season.

Thus it can be seen that for taro and yam there is not enough rain, it is too cold, and thegrowing season is too short. The growing season is barely sufficient for gourd. In the caseof kumara, there is adequate rainfall and a sufficiently long growing season, buttermperatures are about 1° C lower than desirable. It has to be concluded that the climate

UJ J ournal if the R oyal Society if New Z ealand

is d istinctly mar ginal for kumar a and gourd , but these nevertheless must have been th ecrops grown. The sma ll stone mounds are most likely to have been th e plots for gourd,partic ularly as the re is archaeological evidence (B. F. Leach , 1976 : 248) for a centra lpost, wh ich is necessa ry for th e climbing vine. Fern root cou ld have been a minor crop,but soi l and clima te cond itions appear to be generally unfavou rabl e. Bracken fern is onlyra re ly found in botani cal surveys of the area, and is mu ch more common in th e maina lluvial valley of th e Wairar ap a.

Fig. 5 - A reconstruction of an early gardening scene. Gourds would have been grown on thesmall mounds and kumara in the rectangular plots. Some ground has been left fallow. T his scene isbased on site N 168/29 in the Washpool Valley (see Fig. 2).

The changing climate and environment

There is now firm evidence that climatic conditions in New Zealand hav e undergonesignifica nt cha nges during the last millennium (Leach and Leach , 1979: 229 ff.). Thesechanges would have had a profound influence on people and their ac tivities a long th ecoasta l areas of Palli ser Bay for two reasons.

Firstly, at th e best of times, only about 60 days in a summer seaso n of 120 days wouldbe suita ble for co llect ing she llfish in the intertidal area or for inshore fishing (Anderson,197 3: 126). The main prohibiting factor is wind swell whi ch creates very un comfort abl econd itions for foragin g or putting out to sea in a ca noe . H an d line fishing would be a ll butimpossib le on most of th ese da ys of swell , whi ch occur on locally ' fine ' days, wh en coldfront s are affecting weathe r elsewhere. In addition, th e swe ll lasts for some days a fter th epassage of frontal syste ms. In a pe riod characterized by shorter time interv als be twee ncold fronts, th e effects on marin e foraging and fishing mu st have been dram ati c.

Second ly, Palli ser Bay under present climatic condition s is marginal for successfu lkumar a and gourd horti culture . A consistent change in mean annual temperature of 10 Cwould make a conside rable impact on horticultural communit ies, becau se it would beaccompanied by a signi fica ntly shorte r growing season. Suc h a change could well haveturned a critica l situa tion in:o an impossib le one.

Leach - The Prehistory qf the Southern Wairarapa 19

Climatic changes of this order of magnitude are now known to have taken place duringthe period of human occupation, and these must have had far-reaching effects on thecultural and economic history of this region. The sequence of climatic events issummarised in Figure 6.

NZ Glacial Variation NZ Speleothem Curve California D/H Ratio

advance-retreat cold-warm wet-dryAD 2000

. rLillie Ice Age

1f-::d~:a~ t~ ~ • 1~~0-

Li tt Ie Thrust ~Climatic - - - - - - -

Optimum

t ~~~~o_

ADjBc 0

BC1000

BC2000

BC3000

1830

1780

1620

---855---

440

210

620

2780

Fig. 6 - Schematic summary of recent climatic changes in New Zealand.

At the time of first settlement of Palliser Bay, and during the ensuing 450 years, theclimate can be characterized as settled (Leach and Leach, 1979: 234). Many summerswould have been marked by a cycle of northeasterly breezes with drizzle night andmorning and fine days, followed by fresh northwesterly winds with warm daytemperatures, and then brief, cooler southerly winds with only light rain. There is littlelikelihood that more rain fell than at present, but humidity may have been higher. Minordroughts could have been experienced. Wind strengths were probably lower and lessdesiccating. It is possible that anticyclones persisted well into autumn, therebylengthening the growing season by several weeks. Winter conditions would have broughtmore frequent cold fronts with heavier rain and also periods of fine cool weather withsharp inland frosts.

On the other hand, weather in a typical growing season of the period AD 1450 to 1500,and 1600 to 1800, would have been far more unpredictable, with low pressure systemsdominating the weather in both winter and summer. It can be predicted that summer

20 Journal of the Royal Society ofNew Zealand

temperatures would fluctuate between extremes, up to 30° C when northwesterly windsprevailed, and down to 4 or 5° C during the passage of active cold fronts. Shelter wouldprobably become a critical factor for successful gardening. Increased wind strengthduring these periods was undoubtedly reflected in more turbulent sea conditions.

Coupled with these climatic conditions are significant changes in the coastal ecology ofPalliser Bay. There is a complex cause and effect relationship between human economyand the natural environment; unravelling the contribution and precise role of climate andcultural factors in the changes which took place is very difficult. For instance, increasedstormy conditions can result in more tree wind-throw, which in turn can accelerateerosion processes. Man's deforestation of the relatively unstable slopes of the Aorangimountains would have the same effect. On the whole, though, such changes can bedetected in the archaeological record before the onset of climatic deterioration - manwas therefore having a significant impact on the environment. Accelerated shingle fanactivity has been documented at the Pararaki in the 12th and 13th centuries, at TeHumenga by the early 13th century, at Whatarangi after the beginning of the 15thcentury, and at Waiwhero in the 16th century (Leach and Leach, 1979: 236-7). Thesechanges all pre-date the period of worsening climate.

Another line of evidence also attests these environmental changes. Faunal assemblagesfrom archaeological sites dating before about AD 1400 have significant numbers of filter­feeding shellfish, including mussels and tuatua. These species disappear from thearchaeological record after this, and are absent in the modern marine environment. Inaddition, forest birds decline in importance in the faunal assemblages represented inmiddens.

It is suggested that a chain of events occurred with man as the prime initiating force. Itbegan with forest clearance for gardening and was followed by erosion. This raised thegeneral sediment load in streams and rivers, which in turn wiped out sensitive filter­feeding marine shellfish species. With the later poorer climatic conditions, the processgathered momentum, and ultimately produced the now familiar unstable shinglebeaches, braided shingle rivers, and marked hillside erosion and gravel fans.

Subsistence economy and seasonality

The people of Palliser Bay are best described as foraging horticulturalists. Theyexploited the food resources of the rivers and streams, the forest zone, the coastal strip ofland, the intertidal platform, and the offshore area. Of these zones, three stand out as ofgreatest importance in the food quest: the coastal flats were the main areas for kumaragardening, the forest zone was where most birds were caught, and the intertidal andinshore area was where the bulk of fish and shellfish were obtained.

The range of species caught by these people is impressive: at least 50 shellfish species,45 species of bird, 25 of fish, 5 of mammal, and 5 of crustacean. Dominating types ofshellfish were paua, Melagraphia spp., limpets, catseye, and Haustrum haustorium. Themain birds caught were tui, parakeets, pigeons, and shearwaters. Although moa wereeaten, it has been argued that most were obtained in the South Island during tradingexpeditions (B. F. Leach, 1979a: 89-90). We can be quite certain that moa-hunting wasnot a significant aspect of the subsistence economy. Amongst the different kinds of fishtaken, catches were dominated by labrids or spotty, tarakihi, greenbone, red cod,kahawai, eels, and barracouta. Only a few snapper, groper and blue cod were taken,although these are the dominant fish taken today. Very large numbers of crayfish werecaught, and amongst the mammals, only rats were eaten in large numbers. There is noevidence of cannibalism at any period.

Of course, sheer numbers of animals eaten is only part of the story. In terms of bulk offood, the single example of a baleen whale in one of the sites (Smith, 1979: 217) probablyfar outweighs all other remains in all archaeological sites excavated.

Amongst the vegetable foods in the economy, kumara and gourd were the mostimportant, while fernroot may not have been. The evidence of possible fern root attritionplanes on teeth is ambiguous. Firm evidence was obtained for consumption of karaka,

Leach - The Prehistory 0/the Southern Wairarapa 21

hinau and pokaka berries, and no doubt many additional vegetable foods were gatheredtoo.

Many of these foods could not have been obtained in the winter - kumara would havebeen harvested in the autumn (Leach and Leach, 1979d: 261), and the bulk of fishingand shellfish gathering must have been confined to the summer because of turbulent seaconditions in winter. For this reason it is clear that storage of food gathered during thesummer and autumn abundance must have been of great importance if people were tolast out the winter shortages. A portion of the kumara harvest was stored in subterraneanpits, some fish were dried on racks, and birds were potted in fat. Year-round occupationhas been confidently established (B. F. Leach, 1979b).

Settlement pattern

There seem to have been two rather different kinds of settlement pattern in theWairarapa: an early one based on permanent settlement in Palliser Bay, and a laterconfiguration with permanent habitation in the alluvial Ruamahanga valley. Exploitationof these two zones was probably not entirely mutually exclusive, but substantially so. Onthe whole, there is a dearth 0:early archaeological evidence in the main valley, and acorresponding absence of late evidence along the coastal areas.

The early pattern is similar in many ways to that of archaic communities in tropicaleastern Polynesia - the primary focus was the mouth of a river valley. Seven suchcommunities can be identified in Palliser Bay, established within the first 250 years offounding settlement. Coastal and forest resources were of importance to these people, andalthough year-round exploitation has been demonstrated (B. F. Leach, 1979b; Smith,1979: 219), it is suspected that there may have been some shifting in settlementthroughout the year. Two sizeable dwellings have been found in inland locations - onein the Moikau valley from about AD 1180, and another in the Washpool valley fromabout AD 1550. Occupation in at least the summer has been inferred for the former (N.J. Prickett, 1979: 41), and autumn to winter for the latter (B. F. Leach, 1979a: 125). Themain period of activity relating to coastal resources, such as gardening and fishing, wasundoubtedly summer and autumn, and food processing camps and perhaps semi­permanent dwellings were established at the mouths of the valleys. However, in thecourse of the year, people may have retreated inland to escape from stormy winterconditions on the highly exposed coastal flats.

Thus characterized, there are two foci of the early Palliser Bay settlement pattern ­coastal, and inland. Burials in the early period are definitely associated with the coastalarea, and this alone may show the importance of this zone for habitation, since thestandard early Polynesian pattern is for a close association between burials and dwellings(Leach and Leach, 1979c: 210). Houses have not yet been identified archaeologically onthe coast, but traces of several kinds of other domestic structures, such as cooking sheds,have been found (B. F. Leach, 1979a: 81).

It is unlikely that these people in their daily activities ranged much farther than about5 km in any direction, and a community size of 30 to 40 has been suggested (Leach andLeach, 1979d: 266). With this size of group, people must frequently have been obliged toseek marriage partners outside their immediate community, probably amongst groups inother river valleys along the coast.

The late settlement pattern is considerably different. This is essentially related to theNgati-Kahungunu and seems to have been largely based on settlement in the region ofLake Onoke and Lake Wairarapa. Less is known of this configuration archaeologically,and it is reconstructed by extrapolation from historical and traditional sources (Mair,1979). It must be noted that the period in question, late 16th through to 19th century, wasa time of great social upheaval. Thus, there are likely to have been several differentpatterns of settlement. One thing can be stated with confidence, however, it wasmarkedly different to the pattern of previous centuries. People living in the main valleyappear to have been rather more mobile than their predecessors, and fishing partieswould travel from the valley to areas on the coast where they had fishing rights -

22 Journal ofthe Royal Society ()/ New Zealand

distances of up to 30 km (Mair, 1979: 41). There is little evidence of any year-roundhabitation on the coast at this period, and it is more likely that only specialized campswere there. Warfare becomes a significant factor late in the sequence (Mair, 1979: 18),and villages even on flat land were surrounded with stockades. Fortified retreats onhigher ground were also constructed.

Technology and material culture

The early coastal people were not particularly skilled in stone flaking techniques; manyof the adzes which they used had been imported from the South Island in almost finishedform (B. F. Leach, 1979a: 95; Leach and Leach, 1979d: 263). Equally, these people did

l '~ "

:::"',

.....

.',

:.:..';': : '....~':;~" ..:!!.. .. I

~ L:. · . ~ ,

C ~G

em

c==JF

:. I • ~

. ~:. ,.. ~ ... .,

o

oo

"f..

:: <::-;: ;:·~t~~' 1

.~.

o

c~Fig. 7 - Adzes in the southern Wairarapa, made from greywacke.

L each - The Prehistory ofthe SOli them Wairarapa 23

not know how to mak e blades from pr ep ared cores . T heir main spec ia lity was withabrasive ra ther th an percuss ive tech niques, and a single artefact may have involved aseque nce of sca rf-sawing and snapping, filing and gri ndi ng, and finally drillin g. T hesetechniques are mu ch more suited to the ma in rock types availab le locally - greywacke,and unbak ed 'argillite '. These materials were fash ioned into some fine artefacts. Ofparticul ar significance is th e manufacture of greywacke ad zes (see Fig. 7). Some of th esea re clearly copies of th e hogback form , so often encounte red in archa ic New Zealan dsites . In thi s case , however, they were made by what mu st have been a very timeconsuming and difficult hammer d ressi ng technique.

Two othe r tool types - th e schist file and th e greywacke cutte r - are also associatedwith abras ive techniques; both 'ire fairly common in sites of a ll periods . The files ar e wellknown in other parts of New Zealand too, and no doubt served a var iety of functions suchas reaming out the centre of fish~hook tab s. Less well known is the spa ll of greywacke, oneedge of wh ich was used as a cutter with a sawing ac tio n. These tools are similar to somefound in South Cant erbury sites, where th ey are someti mes incor rectly identified as ulu(B. F. Leach , 1979a: 96 ).

Fish ing equipment in Palliser Bay was made from stone, bone and shell. Moa bon e wasth e favourite material , and lures and both one-piece and compos ite hook s weremanufactured. Hooks of a fairly standard style were mad e, but it is noteworthy th at th e'O ruarangi Point ' was in use by about AD 1350 (B. F. Leach , 1979a : 102). Trolling lurepoints whi ch have been ide ntified, a re an unusual form and appear to be a region al style(Ande rson, 1979: 57 , B. F. Leach , 1979a: 103 ).

Also with a restricted d istribution is a type of tri angular sec tioned stone lur e shankwh ich has two distal knob s on th e ventral surface (N. J. Pri ckett , 1979: Fig. 5; B. F.

(f{\ rr \l~~K L

__~_~ cm

Fig. 8 - One-piece shell fish-hooks from Palliser Bay.

24 Journal if the Royal Society if New Zealand

Leach, 1979a: Fig. 15). This form is confined to the Cook Strait region (B. F. Leach,1979a: 100-101), occurring on both islands. Other types of stone lure shanks are alsocoeval with this form in Palliser Bay, but one type does apparently have chronologicalsignificance. This is an oval or triangular sectioned stone lure shank with a laterallynotched snooding attachment at both ends. Only two have been found; the one which iscomplete is unperforated, but the incomplete specimen could have been dorso-ventrallydrilled. These two lures are clearly similar to those of the Coromandel-Hauraki Gulf area(N.]. Prickett, 1979a: Fig. 5; B. F. Leach, 1979a: 100). They occur in one of the earliestarchaeological sites in Palliser Bay and could be a positive indication of the source of thefounding group. However, the Cook Strait lure type was found closely associated and alsooccurs in the later sites, whereas the former does not.

Shell fish-hooks are invariably one-piece in form (Anderson, 1979: Fig. 3; B. F. Leach,1979a: Fig. 21), and seem to be confined to contexts earlier than the 14th century. Someexamples are illustrated in Figure 8:

Bird spears are reasonably common in the archaeological sites (Fig. 9) - a sure sign ofthe importance of birding in the economy. There are several distinctive forms (B. F.Leach, 1979a: 104 ff.).

Rarer artefacts are stone food pounders, amulets, gaming stones, sandstone grinders,pumice net floats, stone sinkers, and bone tattooing chisels. The latter are represented bytwo specimens: one from about AD 1180, and the other from about AD 1350. Both arenotched at the proximal end (B. F. Leach, 1979a: Fig. 2; 107 ff.), showing that this form,so common in later archaeological sites, has a respectable antiquity in New Zealand. It iscontemporary with the proximal drilled variety found at Wairau Bar, but is morecommon in early sites in the Hauraki Gulf and Northland areas.

0'o

TYPEG•o c::::>TYPE E

•=

;,1_11 ..

I.

TYPE 0

J

=

AeN

r~

-ii~~.''~

e>G

I:

FQ

c-"" D\J E"" =

TyPE C

o = Eo

TYPE A

co

~

•o

.'oTYPE B

Fig. 9 - Bone harpoons from Palliser Bay.

Leach - The Prehistory if the Southern Wairarapa 25

Trade and exchange

Both economic and ceremonial exchange were of fundamental importance in thestructure of pre-European New Zealand society. An enormous range of material goodsflowed hither and yon and was the concrete expression for the transfer of equallyimportant abstract social values such as status. Mauss (1966: 9) has described thisspiritual transfer as the LeitmotifofMaori custom and suggested that it could be positivelydangerous to keep the concrete expression of the gift. A material gift is itself a strongmotivating force for ever further circulation of property and status (B. F. Leach, 1978a:391 ff).

People in the southern Wairarapa were involved in a regular system of exchange andcommunication with other groups in New Zealand from the period of first settlement ofthe region. This is evident in the range and quantity of imported material goods in thearchaeological sites. The main surviving residues of this exchange network are differentkinds of stone, but we can be certain that many other more perishable items wereinvolved too. Chert was being obtained from the Tora area; limestone from White-Rock,Tora, and Akitio in Hawkes Bay; and obsidian from Northland, Coromandel, Bay ofPlenty, Rotorua, and Taupo. From the South Island, argillite was obtained fromD'Urville Island, nephrite from the West Coast, and silcrete all the way from CentralOtago. Thus, stone materials alone were coming into Palliser Bay from as far away as850 km to the north and 750 km to the south. It IS also believed that moa bone was anitem of trade, as the main species in the sites is a South Island variety (B. F. Leach 1979a:89-90). In addition, numerous tui and parakeets caught locally were preserved in aspecial manner (B. F. Leach, 1979c), and were probably just one of the items beingtraded out of the region. In the later period, historical records show that eels from LakeOnoke were traded throughout the North Island. Mair has reconstructed an impressivelist of items ranging from kumara, mats and canoes, to feathers and scented oil, whichwere flowing north and south, through the hands of southern Wairarapa people (Mair,1979: 22-23).

It is possible that these groups in the Wairarapa should be seen as entrepreneurs ­able to make a profit from handing on some northern obsidian to people in the SouthIsland and at the same time profiting from handing on to northern people argillite adzeblanks made in the South Island. Ifthey were middlemen in this fashion they would be ina good position to siphon off for their own use a portion of the goods flowing in bothdirections. It is not known how much long distance transportation was involved, incontrast to neighbour-to-neighbour exchange. Given the more sedentary settlementpattern suggested for the early period compared with that of later communities, it couldbe that there was a tendency for long distance transportation to supplant an earlier systemof essentially short distance transfer.

The extent to which early communities in Palliser Bay relied on imported stone isremarkable. In the course of time, the proportion of imported objects as against thosemade of local stone rose from 77% (AD 1150) to 83% (AD 1350) to 94% (AD 1550) (B.F. Leach, 1978a: 397).

This shows not only that the desire for imported materials was always high, but thatcommunication channels with places elsewhere in New Zealand were very strong. It isinteresting that these external links changed in character through time. The great heydayof trade was clearly in the middle prehistoric period (about AD 1350) and slowly shrankto a closer involvement with people closer to home, particularly around the Cook Straitregion.

THE PREHISTORIC SEQUENCEPeriod I: Successful colonization (AD 1000 to AD 1250)

There is firm evidence from several parts of Palliser Bay that a number ofvalley-basedcommunities were established by about AD 1050 to 1200, and it is likely that the firstsettlers reached the area not much before this - perhaps by AD 1000 (Fig. 10). Thepeople appear to have had close links with groups farther to the north (from where they

26 J ournal cf the Royal Society ofNew Zealand

2000

1900

1800

"ac-, ' C ~

"i: c8 ~~' i: '- ...!.:I:

'!!II i~fL~!1 ":"& z ",-"_ <{ co~~ --J ';!Z..::: Vl

} T~cu ..... ~1600 g~::::> ~

~ ~ ]a ~ E

0:: ~~

"a E

~£

1500

Z :,':;::~. ~

:- w. J:

jo _~=====-:

.. . .: ..-:

jl

"".."I"~ ~

" '"., '"::: u "0..J ~ <

iii .!

o ~c uo "

l' II" .'" w

;; ~

> ".'.,,1:~ ....

VIIIEuropeanHiatorle

~,~ob 1840

VIEul1

Protohtatorl c

i -'ffl'V

N••Coloni a.t loD

-----+,~,IV

Abaodml!lUllat

I1550

1400

I I ~~~11

RotriD...I

1'1COQ8OUdaUOD

I L"00

~ 250

e.§0.

t--..,,~ 1200° 1u

§u

Il1>I

Bucc • • • tuJ

~ .... .. .Coloota.tioD

1100

IIIEooaomlc..

Eu"f1f'ODmentalDeoIl D.

I I ·1450

Titoki Pit

\o!:l,hpa>llo<>lls &l.,.11IITerract fob..tsf

Titokl Pit

H()lkau House

Washpool Level II

Ptnd HIdden

PararOk l Walls( ru c, nt I'1 lCidt nKo...okawa Wallsf'1ClJnd GardenParoral'll 'Wat ls

SInalI Hidden

Ra.nj Pit

r.=~"~I\'81oclo; MiddenPororciu Wal ls

ParonJk, 'Walls

Whatorongl 'Wall 50Black Rock s Walls

Garden Terrac.

(I. ftWJlwhero Wblls

I ! 1000 1000 AD

Fig. 10 - Schematic sequence of the prehistory of the southern Wairarapa.

Leach - The Prehistory if the Southern Wairarapa 27

may have originated), such as Coromandel and Bay of Plenty, as obsidian from thesesources dominates. However, some obsidian also derives from sources around Rotoruaand Taupo in this period, suggesting the possibility of communication with central NorthIsland people also. Some of the earliest stone lures are fashioned in a style common in theCoromandel and Hauraki Gulf area, which again may point to a northern origin for thefounder population.

It is clear that people engaged in considerable experimentation with local stone - andwere attempting to apply old skills to a new lithic environment. The main rock type foundis greywacke, and this is not easy to fashion into the usual types of stone adze which thesepeople were familiar with, such as the hogback and lA forms. Nevertheless, theycertainly attempted to do so.

This period was in the Little Climatic Optimum, when seas were somewhat calmerthan they are today, and the growing season was longer by perhaps four weeks. Kumaraand gourd were successfully cultivated, and areas of land were neatly laid out intorotational plots - some under cultivation, others lying fallow. Bush birds weresystematically caught, and some parts of them preserved in fat for trade or winterconsumption when poor weather made conditions unpleasant enough for people to stay athome.

The village was usually centred on the mouth of a river valley, and the whare-puni canbe clearly identified at this period as a carefully laid out and constructed house, verysimilar to those seen in New Zealand in the 19th century. Small cooking sheds wereestablished on the coast, and people were buried with suitable ceremonial ritual nearthese and other domestic structures. This is a continuation of the standard practice oftropical eastern Polynesia. It appears that although these people had a hard and activelife, they had a nutritionally adequate diet, and led a moderately healthy existence.

Dense forest conditions were nearby, with quiet rivers flowing through bush cladvalleys to the sea. Mussels and tuatua were available in especially stable pockets ofbeaches along the coast.

Period II: Regional consolidation (AD 1250 to AD 1450)The climax population of Palliser Bay has been estimated at about 300 (H. M. Leach,

1976: 182) given the kind of economic system which prevailed. This was spread overseven village based communities centred on the mouths of the major river valleys of thearea. Individual group sizes may have varied somewhat, and on the basis of shellfishgathering behaviour Anderson (1973: 127) has suggested parties of 10 to 25 people atBlack Rocks, while H. Leach suggests communities of between 30 and 40 on the basis ofareas under cultivation at anyone time (1976: Ch. 9; see also Leach and Leach, 1979d:266). A founding community of perhaps 25 people based at one such river valley wouldreach the climax population of 300 in about 250 years (assuming a 1% growth rate perannum). We can be reasonably confident that by about AD 1250, the carrying capacity ofthe land along the shores of Palliser Bay had been reached.

It is in the ensuing 200 years that trading contacts seem to have been most highlydeveloped, and this is evident in the rich variety of imported stone types (B. F. Leach,1978a: 395, 403; Leach and Anderson, 1978). Obsidian, for example, was enteringPalliser Bay from no less than seven different geological sources in the North Island ­Huruiki, Purangi, Cooks Bay, Mayor Island, Rotorua, Ongaroto, and Taupo. More to thepoint, links between communities in Cook Strait firmed up at the expense ofthose furtherafield (B. F. Leach, 1978a: 401, 403; K. Prickett, 1979: 179), and this is an indication ofregional consolidation. This is especially evident in the increased importation of chertfrom the Tora area, and metasomatised argillite from D'Urville Island (B. F. Leach,1978a: Table 1).

During this period it is suggested that there was reasonable economic stability. Therewere still beds of such filter-feeding shellfish as mussels and tuatua to be exploited. Thisis documented in the Crescent Midden at Black Rocks (Anderson, 1979: 53), and LevelII at the Washpool (B. F. Leach, 1979a: 88), and suggests that coastal erosion processes

28 Journal if the Royal Society if New Zealand

had not yet begun to accelerate. At the same time, the number of forest birds which werebeing caught definitely decreased (Leach, 1979c: 108; Anderson 1979: 55), which nodoubt reflects the significant encroachment of gardens into nearby forested areas. Datedarchaeological sites show that by this time most of the coastal strip in Palliser Bay wasbeing gardened.

Period III: Economic and environmental decline (AD 1450 to AD 1550)

The onset of this period was probably associated with a brief climatic recession knownin the northern hemisphere as the 'Medieval Thrust'. Filter-feeding shellfish disappearfrom the archaeological record suggesting unstable nearby beaches brought about byhigh sediment loads in the rivers, which were carrying significant quantities of slopeerosion products as a result of deforestation. Settlements were certainly up to 2 km inlandup the river valleys, and people buried their dead in hide-away locations - a sure sign ofgrowing unrest. The lines of supply of imported stone by this time, while still outside theWairarapa, had shrunk to areas rather closer to home, and the people were in a state ofnear-dependence on imported raw materials (8. F. Leach, 1978a: 397).

Above all, the evidence from the Cleft Burial site in the Washpool shows a dramaticincidence of Harris lines in limb bones - the result of adverse economic conditionswhich cause periodic malnutrition or near starvation.

Period IV: Abandonment (AD 1550 to AD 1625)

Under such circumstances of economic poverty and worsening climate and seaconditions, the scene was set for eventual depopulation of the area. Exactly when thisprocess began is not known but several lines of evidence suggest the dates given. The'Little Ice Age' began around AD 1600, and there are strong oral traditions of a finalexodus of the Rangitane in AD 1625. The land snail fauna from the Pond Midden atBlack Rocks, dated to about AD 1750, shows that by this time a scrubland had developedover the coastal platform - such as would follow a period of abandonment (Wallace,1979). The later re-establishment of shellfish beds at Black Rocks confirms the absence ofpeople engaged in marine foraging (Leach and Anderson, 1979: 161). It is suggested thatthis movement away from Palliser Bay may have begun as early as AD 1550, around theend of known habitation of the Washpool valley.

Period V: New colonization (AD 1625 to AD 1769)

The earliest migrations ofthe Ngati-Kahungunu probably occurred in the first half ofthe 16th century when a group left Poverty Bay for Heretaunga. It appears that they livedthere until AD 1625, when they moved on to Southern Wairarapa (Mair, 1979: 12;Bagnall, 1976: 5). This period, which lasted until Captain Cook arrived in New Zealandin 1769, is practically unknown archaeologically, but what little evidence is found in theregion is interesting. The Pond Midden at Black Rocks has a more restricted faunalassemblage than earlier sites and is very likely to be the result of a fishing party ofthe kindknown during the protohistoric period (Mair, 1979: 20). At that time, people in the mainalluvial valley of the Wairarapa made short expeditions to areas on the coast where theyhad fishing rights. This interpretation of periodic non-intensive exploitation of the coastis reinforced by the general size of species caught in the Pond Midden (Anderson, 1979:62); and non-residence is suggested by the above-mentioned dense scrubland conditionsnearby.

Thus characterized, this period is one of absentee ownership of the coastal areas ofPalliser Bay. This then was probably the time at which significant habitation ofthe mainalluvial valley and river and lakes district began. There are only indirect and isolatedindications of earlier settlement in this zone.

In the absence of excavations in this area, the subsistence base, of what was essentiallyan inland community, can only be guessed at. However, it must be remembered that thisperiod was in the middle of the 'Little Ice Age', when cultivation of all known Pacific

Leach - The Prehistory cf the Southern Wairarapa 29

Island crops in this area would have been extremely difficult. Of profound importance tolater groups, where records are clearer, were the eels of the lower lakes district. Periodicclosure of the sand bar at Lake Onoke facilitated enormous catches of eels, estimated at10 to 30 tons (Mair, 1979: 22). These were highly prized throughout the North Island intrading exchanges. Resident groups, even in this period of rather poorer climate,therefore, may have had a secure basis to their economy by preserving eels for the winter,and exchanging some for kumara grown farther to the north. Fernroot too is known tohave been an important resource in this area in later periods and could have beenencouraged or semi-cultivated. The area of the Ruamahanga River valley was denselyforested, and would have provided many foods such as birds, rats, and berries.

Period VI: Early protohistoric (AD 1769 to AD 1840)

Captain Cook arrived in Palliser Bay on February 9, 1770. When people in threecanoes paddled up to the Endeavour off the east coast just to the north of Cape Palliser,they immediately asked for nails! Cook observed that they had never seen them beforebut evidently knew how to use them. He surmised that they had learned about nails fromthe people of Cape Turnagain whom he had visited the previous year (Mair, 1979: 11).Thus, the protohistoric can be reasonably held to have begun with Cook's first contactwith the Maori farther north in 1769.

This was a period of great upheaval in the Wairarapa, induced in the main by theactivities of Te Rauparaha, in his southward drive along the west coast and into CookStrait. The Wellington and Wairarapa districts were visited by war parties of Ngapuhi,Ngati Whatua and Ngati Toa in 1820 and later by the Amiowhenua Expedition of1821-1822. Groups on the west coast were supplanted by Taranaki tribes, and someretreated into the mountains or over them into the Wairarapa. In 1825-26 migrants fromTaranaki also entered the Wairarapa valley (Mair, 1979: 13).

This period of social unrest and skirmishes resulted in the gradual and thenaccelerated exodus of people from the Wairarapa to the Mahia peninsula area. By 1835there appear to have been no more than about 100 people resident in the region, from aclimax population of around 900 (Mair, 1972: 222; 1979: 15). No doubt, many of thefortified pa sites in the lower valley were built during this period of conflict.

Relatively poor climatic conditions still prevailed in this period, and there would havebeen a continuation of the non-horticultural foraging economy. At the same time,significant changes would have come about as a result of the spread of European foodresources. By 1839, for example, the hunting of wild pigs is recorded (Mair, 1979:20).The trade of fish and crayfish is consistently mentioned in historical accounts.

Period VII: Late protohistoric (AD 1840 to AD 1853)

The signing of the Treaty of Waitangi in 1840 was the turning point in the eventualresettlement of the Wairarapa. It was believed that land rights reverted to the situationimmediately before the mass migration; secure in this knowledge, groups began to re­occupy their old lands. The Heretaunga hapuleft on one day, sailing home in 69 canoes.The various hapu at Te Mahia were not as quick to return, but most had left by 1842. Apopulation of 604 was recorded in the southern Wairarapa in Kemp's census in 1848-49(Mair, 1972: 224; 1979: 15).

The Treaty of 1840 marks the beginning ofthe historic era for most ofthe Wellingtonarea, but not so for the Wairarapa. This region was not explored by Europeans until1841, and squatters did not move into the valley until 1844. It was not formally settled byEuropeans until a government-approved scheme began in 1853 (Mair, 1979: 12).

European crops were extensively cultivated in the late protohistoric period. Cabbageand turnips grew profusely in the wild, and pumpkins, melons, corn, and wheat werecultivated. Kemp records Wairarapa Maori growing 15 acres of wheat, 32 acres of corn,102 acres of potatoes, and 121/2 acres of other garden produce in 1850 (Mair, 1979: 21).As before, the great eel harvest of the Lake Onoke area was of continuing importance.

30 Journal ofthe Royal Society ofNew Zealand

DISCUSSION

One of the most interesting features of the archaeological record in the southernWairarapa is that there is no evidence of a marked change from Archaic to Classicculture, often postulated in New Zealand prehistory. The threads of change can bedetected in many aspects of the prehistoric culture and environment, but on the wholethese changes were gradual, and explicable rather than sudden and dramatic. They alsotook place at different times and different rates over a long period rather than all at once.This is not to suggest that there were no significant events in the prehistoric period ­quite the contrary, as the foregoing has shown. It seems, though, that the archaeologicalevidence does not easily lend itself to polarisation into the two separate systems ­Archaic and Classic - which bedevil reconstructions in other parts of New Zealand. Inthe southern Wairarapa, the human occupation and its effects on the environment inseveral distinct periods have been characterized by a blend of economic, environmentaland social features.

Some concluding remarks are appropriate about the possible role of the southernWairarapa people in the emergence of what has been called 'Classic Maori Culture' inthe South Island. The question of the source of this influence in the South Island is ofconsiderable importance and has been much debated.

A study of oral traditions on this subject has shown that there were at least threemovements of people which are subsumed under the general name of 'Ngai-Tahumigration, (B. F. Leach, 1978b: 19).

(i) Ngai-tahu group, headed by Tuteahunga, from Hataitai in Wellington, aboutAD 1635.

(ii) Ngati-kuri group, headed by Manawa, from Hataitai in Wellington, aboutAD 1685.

(iii) Ngai-tuahuriri group, headed by Turakautahi, from Turanga[nui], in Palliser Bay,about AD 1710.

The leaders of these three movements were all descendants of Rakawakakura, a wellknown Ngati-Kahungunu figure (B. F. Leach, 1978b: 29). Tuahuriri, the eponymousancestor of the third movement, was the father of Turakautahi. He made severalexpeditions to the South Island, one with a group of 70, and another with 170 people.Oral traditions suggest that most of these people returned to Palliser Bay (B. F. Leach,1978b: 21). Such excursions sound more like trading expeditions than expansionistvoyages. Tuahuriri himself was drowned in Cook Strait when his canoe overturned onone of these trips. Just how many people went with his son, Turakautahi, to the SouthIsland is not known; but on the whole, the impact of the Wellington derived groups of theNgati-Kahungunu led by Manawa and Tuteahunga may well have been greater, sincethey figure more prominently in South Island traditions.

Although there are oral traditions suggesting a close relationship between the Ngati­Kahungunu and those speaking the southern dialect (B. F. Leach, 1978b: 20), untilrecently there seemed little hard linguistic evidence of this relationship. This difficultyhas been rectified by Harlow's recent research into New Zealand dialects, which clearlyshows that the southern dialect is closest to those of the Ngati-Kahungunu and Tuhoe(Harlow, 1979). This result supports the traditional Ngati-Kahungunu affiliations of theleaders of the three expeditions mentioned above but does not necessarily strengthen thecase for any supposed mass movement from the southern Wairarapa to the South Island.

The South Island Ngai-Tahu were renowned for their skills in the manufacture ofnephrite tools using abrasive techniques. It has been argued (Leach and Leach, 1979d:264) that this type of technology could have been imported into the South Island bygroups from the Wairarapa, who had a long established tradition of stone working byabrasive techniques. This type of technology, of course, was established in the southernWairarapa long before the arrival there of the Ngati-Kahungunu, and could have beenintroduced to the South Island by earlier migrants, such as the Rangitane.

It must be noted that the movement of a group of people under the direction ofTurakautahi from the shores of Palliser Bay to the South Island about AD 1710 was not

Leach - The Prehistory of the Southern Wairarapa 31

an isolated phenomenon. The Tini-o-awa, who entered the Wairarapa as a result of theexpansionist activities of the Ngati-Kahungunu farther north at an unknown period,lived for a time alongside the principal occupiers of the Wairarapa, the Rangitane, butultimately emigrated to the South Island (Mair, 1979: 12). The Rangitane too, confrontedwith the incoming Ngati-Kahungunu, moved to the South Island. This event is wellknown in oral tradition and apparently followed a peaceful encounter that took placeabout AD 1625. The Rangitane exchanged their lands for canoes and weapons suppliedby the Ngati-Kahungunu, Thus, it appears that there may have been a long series ofcontacts and movements between the southern Wairarapa and the South Island, ratherthan a single decisive 'migration'.

The encounter between the Rangitane and the Ngati-Kahungunu occurred at the endof what is believed to have been a period of progressive abandonment of the southernWairarapa, as an area of little further economic viability, given the kind of economicsystem that had prevailed there for the previous 500 years. Thus, the Rangitane may wellhave considered that they got the best of this trading encounter with the Ngati­Kahungunu. Perhaps this is what their leader, Te Rerewa, implied in his ambiguousparting remark to the new owners from the vantage of his departing canoe:

You watch the whawhakou [Eugenia maire] tree; if it bear much fruit it will be ayear of famine, no birds, no rats; that is a sign of the Wairarapa district - andwhen you see the waters of the lake flow to its source it will be a year of plenty(Bagnall, 1976: 6).

Whether the lake to its source could flow, before Great Birnam wood to high Dunsinanehill could go, is a matter for sober reflection. The prosaic interpretation of this remark isthat when the gravel bar at Lake Onoke closes, the waters back up and will beaccompanied by a great eel harvest (Bagnall, 1976: 6). However, the imagery is morecomplex - Te Rerewa could have been making a poetic remark on the difficulties histribe had experienced living in the southern Wairarapa.

In the foregoing it must be abundantly clear that a view of prehistoric man as a 'noblesavage' in harmony with nature, and an 'archetypal conservationist' ofthe environment isnot one which can be sustained, at least for this region of New Zealand. The combinedeffects of forest clearance and over-exploitation of marine foods ultimately led to suchpronounced erosion and depletion of natural resources that further occupation of thecoast, even with introduced crops and animals, was impossible. The foothold which thesepeople had was always tenuous, since kumara horticulture was marginal at the best oftimes. The onset of deteriorating climatic and sea conditions must have been the laststraw. One must be careful, though, in interpreting the human responses to this chain ofevents, as Anderson (1979: 64) has pointed out:

... it would be invidious to draw the conclusion that the . . . inhabitants wereignorant of, or unsympathetic to, the need for ... conservation; it may well havebeen a luxury they could simply not afford.

REFERENCESAdkin, G. L. 1955. Archaeological Evidence of Former Native Occupation of Eastern Palliser

Bay. Journal ofthe Polynesian Society 64(4): 450-480.Anderson, A. J. 1973. A rchaeology and behaviour. M.A. thesis, University of Otago: Dunedin.

1979. Prehistoric Exploitation of marine resources at Black Rocks Point, PalliserBay. In Leach and Leach, 1979a. Pp.49-65.

Bagnall, A. G. 1976. Wairarapa: an historical excursion. Hedley's Bookshop.Barrow, T. 1959. An Archaic type of Maori hei-tiki from the Wairarapa East Coast. NZ

A rchaeologicalAssociation Newsletter.' 2(4): 6-7.Cairns, K. R. 1959a. A Hangi site at Glenburn. NZ Archaeological Association Newsletter2(4):

26.1959b.1960.1961.

Work in the Wairarapa. NZ Archaeological Association Newsletter 2(4): 19-20.Fieldwork, Wairarapa. NZ Archaeological Association Newsletter 3(4): 24.Wairarapa District. N Z A rchaeologicalAssociation Newsletter 4(4): 22-24.

32 Journal ofthe Royal Society qf New Zealand

1971. Rediscovering the Wairarapa. Wairarapa Times Age February 19, p. 14.

Davis, S. 1957. Evidence of Maori occupation in the Castlepoint Area. Journal of thePolynesian Society 66: 199-203.

1959. A summary of field archaeology from the Dominion Museum group. NZA rchaeologicalAssociation Newsletter 2(4): 15-19.

Harlow, R B. 1979. Regional Variations in Maori. NZ Journal ofArchaeology 1: 123-138.

Hitchings, M., and M. O'Rourke. n.d. Site Survey Fieldbook: Eastern Wairarapa.Manuscript.

Houghton, P. 1975. The people of Wairau Bar. Records of the Canterbury Museum 9(3):231-246.

1980. The First New Zealanders. Hodder and Stoughton.

Leach, B. F. 1976. Prehistoric communities in Palliser Bay, New Zealand. Ph.D. thesis.University of Otago: Dunedin.

--- 1978a. Four Centuries of community interaction and trade in Cook Strait, NewZealand. In J. Specht and J. P. White (Eds.), Trade and Exchange in Oceanial and AustraliaMankind 11(3). Pp. 391-405.

1978b. The Ngai-tahu migration: the Norman Conquest of the South Island. NZA rchaeologicalAssociation Newsletter 21(1): 13-30.

1979a. Excavations in the Washpool Valley, Palliser Bay. In Leach and Leach,1979a. Pp. 67-136.

1979b. Fish and crayfish from the Washpool Midden site, New Zealand: their use indetermining season of occupation and prehistoric fishing methods. Journal rifArchaeologicalScience 6(2): 109-126.

1979c. Maximizing Minimum numbers: avian remains from the Washpool Middensite. In AJ. Anderson (Ed.), Birds of a Feather. British Archaeological Reports. Pp. 103-121.

Leach, B. F., and A J. Anderson. 1974. The transformation from an estuarine to lacustrineenvironment in the Lower Wairarapa.Journal of the Royal Society ofNew Zealand 4(3): 267-275.

1978. The prehistoric sources of Palliser Bay obsidian. Journal ofA rchaeological Science5: 301-307.

1979. Prehistoric exploitation of crayfish in New Zealand. In A J. Anderson (Ed.),Birds rifa Feather. British Archaeological Reports. Pp. 141-164.

Leach, B. F., and H. M. Leach (Eds.). 1979a. Prehistoric Man in Palliser Bay. NationalMuseum of New Zealand, Bulletin 21.

1979b. The Wairarapa Archaeological Research Programme. In Leach and Leach,1979a. Pp. 67 -136.

1979c. Burial positions and orientations in Palliser Bay. In Leach and Leach,1979a. Pp. 205-213.

1979d. Prehistoric Communities in Palliser Bay. In Leach and Leach, 1979a.Pp.251-272.

Leach, H. M. 1976. Horticulture in prehistoric New Zealand. Ph.D. thesis. University ofOtago: Dunedin.

1979a. Evidence of prehistoric gardens in eastern Palliser Bay. In Leach and Leach,1979a. Pp. 137-161.

1979b. The significance of horticulture in Palliser Bay for New Zealandprehistory. In Leach and Leach, 1979a. Pp. 241-249.

Leach, H. M., and B. F. Leach. 1979. Environmental change in Palliser Bay. In Leach andLeach 1979a. Pp. 229-240.

Mair, G. M. 1972. The protohistoric period of Wairarapa culture history. M.A thesis.University of Otago: Dunedin.

1979. Maori occupation in the Wairarapa during the protohistoric period. In Leachand Leach, 1979a. Pp. 11-28.

Mauss, M. 1966. The Gift. Cohen and West.

Mitcalfe, B. 1968a. Flat Point - Glenburn Site Survey. Manuscript, Wellington TeachersCollege.

Leach - The Prehistory ofthe Southern Wairarapa 33

1968b. White Rock - Te Awaite Site Survey. Manuscript, Wellington TeachersCollege.

1968c. White Rock Site Survey. Manuscript, Wellington Teachers College.

1968d. Castlepoint - Mataikona: Preliminary Field Report. Manuscript, WellingtonTeachers College.

1968e. Wairarapa Coastal Survey: Vegetation.

1968f. The Wairarapa Coastline: one hundred miles and a thousand years away. TeKaunihera Maori, Spring 1968: 25-33.

1969. Towards a regional definition of Maori New Zealand: coastal Wairarapa.Manuscript, Wellington Teachers College.

Palmer, J. B. 1961. Some aspects of New Zealand field archaeology. Journal rif the PolynesianSociety 70: 466-470.

Park, G. N. 1970. A Maori oven site in the Tararua range, New Zealand: apaleoclimatological evaluation. NZ Archaeological Association Newsletter 13(4): 191-197.

Prickett, K. 1979. The stone resources of early communities in Palliser Bay. In Leach andLeach, 1979a. Pp. 163-184.

Prickett, N. J. 1979. Prehistoric occupation in the Moikau Valley, Palliser Bay. In Leachand Leach, 1979a. Pp. 29-47.

Scarlett, R. J. 1960. Conference Report 1960. NZ Archaeological Association Newsletter 3(3):3-11.

Smart, C. D. 1966. The ditch -and-bank fence. N Z A rchaeological Association Newsletter 9:19-28.

Smith, 1. W. G. 1979. Prehistoric sea mammal hunting in Palliser Bay. In Leach and Leach,1979a. Pp. 215-224.

Sutton, D. G. 1974. Resurrection of the prehistoric dead. M.A. thesis. University of Otago:Dunedin.

1979. The prehistoric people of Eastern Palliser Bay. In Leach and Leach, 1979a.Pp. 185-203.

Wallace, R 1979. Land snails from archaeological sites in Palliser Bay. In Leach and Leach1979a. Pp. 225-228.

Wellman, H. W. 1962. Holocene of the North Island of New Zealand, a coastalreconnaissance. Transactions of the Royal Society rif New Zealand (Geology) 1: 29-100.

Received, 25 September, 1980